Asthma and Invasive Pneumococcal Disease
http://www.100md.com
《新英格兰医药杂志》
To the Editor: Talbot et al.1 (May 19 issue) find that in the absence of coexisting conditions such as diabetes, alcoholism, or infection with HIV, which are known to confer a high risk of invasive pneumococcal disease, the excess presence of invasive pneumococcal disease among persons with asthma is one to three episodes per 10,000 person-years. The presence of coexisting conditions was determined from hospital and pharmacy records for the year before the index date, defined as the date of isolation of Streptococcus pneumoniae. However, HIV infection often goes undetected, even in patients receiving medical care,2 until a known HIV-related illness develops, such as invasive pneumococcal disease.3 Invasive pneumococcal disease has been estimated to be as much as 100 times more common among persons with HIV infection than in the general population.4,5 It has been recommended that a diagnosis of invasive pneumococcal disease trigger testing for HIV antibody.3,5 The restriction of coexisting conditions only to those documented on or before the date of isolation of S. pneumoniae may have resulted in substantial underestimation of the incidence of HIV infection. The effect of such an underestimation on the study findings should be considered.
Deborah Cotton, M.D., M.P.H.
Boston University School of Medicine
Boston, MA 02118
cottond@bu.edu
References
Talbot TR, Hartert TV, Mitchel E, et al. Asthma as a risk factor for invasive pneumococcal disease. N Engl J Med 2005;352:2082-2090.
Kuo AM, Haukoos JS, Witt MD, Babaie ML, Lewis RJ. Recognition of undiagnosed HIV infection: an evaluation of missed opportunities in a predominately urban minority population. AIDS Patient Care STDS 2005;19:239-246.
Garcia-Leoni ME, Moreno S, Rodeno P, Cercenado E, Vicente T, Bouza E. Pneumococcal pneumonia in adult hospitalized patients infected with the human immunodeficiency virus. Arch Intern Med 1992;152:1808-1812.
Redd SC, Rutherford GW III, Sande MA, et al. The role of human immunodeficiency virus infection in pneumococcal bacteremia in San Francisco residents. J Infect Dis 1990;162:1012-1017.
Nuorti JP, Butler JC, Gelling L, Kool JL, Reingold AL, Vugia DJ. Epidemiologic relation between HIV and invasive pneumococcal disease in San Francisco County, California. Ann Intern Med 2000;132:182-190.
To the Editor: Talbot et al. use a case definition of asthma that relies on medical-record review of the diagnostic codes entered or medication-prescription records, or both. Physical-examination data, pulmonary-function assessment, and laboratory information are not incorporated into the case definition. The decision to use a diagnosis of asthma by a noninvestigator physician as the study case definition may result in the misclassification of study subjects. The authors cite two published reports to support the validity of their choice of case definition.1,2 However, the current investigation does not, in fact, use the same case definition as was used in those studies. Moreover, the cited investigations studied patients with asthma who were over the age of 65 years, whereas the average age in the current study population is 28 years and persons older than 49 years were excluded. The authors should acknowledge the possibility of misclassification of study subjects as a potential limitation of their analysis and cite reports that validate their case definition.
Ware G. Kuschner, M.D.
Veterans Affairs Palo Alto Health Care System
Palo Alto, CA 94304
kuschner@stanford.edu
Robert A. Kuschner, M.D.
906 N. Wayne St.
Arlington, VA 22201
References
Hartert TV, Togias A, Mellen BG, Mitchel EF, Snowden MS, Griffin MR. Underutilization of controller and rescue medications among older adults with asthma requiring hospital care. J Am Geriatr Soc 2000;48:651-657.
Hartert TV, Speroff T, Togias A, et al. Risk factors for recurrent asthma hospital visits and death among a population of indigent older adults with asthma. Ann Allergy Asthma Immunol 2002;89:467-473.
The authors reply: We appreciate the comments of Dr. Cotton regarding undiagnosed HIV infection as a risk factor for invasive pneumococcal disease. We reexamined the Tennessee Medicaid program (TennCare) records of cases of invasive pneumococcal disease for the years after patients had received a diagnosis of invasive pneumococcal disease for health care visits that recorded a coded HIV diagnosis or a prescription for HIV-specific medications. In 9.8 percent of the patients coded as having invasive pneumococcal disease (34 of 347) and 0.4 percent of the controls without a diagnosis coded as HIV or other risk factors for invasive pneumococcal disease (22 of 5541), patients were later given a diagnosis of HIV infection. Of patients with invasive pneumococcal disease who had asthma, 5.9 percent (3 of 51) were later given a diagnosis of HIV infection, as compared with 10.5 percent of those without asthma (31 of 296). We agree with Dr. Cotton that HIV testing should be strongly considered for persons who present with invasive pneumococcal disease.
Drs. Kuschner and Kuschner appropriately noted the limitations of the use of administrative data and questioned our definition of asthma. A diagnosis of asthma is established on the basis of the patient's history; documentation of reversible airflow obstruction by spirometry, methacholine challenge or another challenge test, or both; and the exclusion of any alternative diagnoses.1 Previous studies using Medicaid data revealed that diagnostic coding data have an excellent predictive value for asthma diagnosed with the use of these criteria.2,3 The confirmation of the use of diagnostic codes in a younger cohort (15 to 44 years of age) has also been published.4
Those earlier studies also made it possible to assess the validity of our diagnostic algorithm for asthma, which included the use of asthma-specific medications, on the basis of standardized criteria and chart review conducted in an older adult cohort (65 years of age) (unpublished data). This was a more sensitive, though less specific, definition in the older group, since it also identified subjects with chronic obstructive pulmonary disease (COPD). These findings provided the rationale for limiting our study to those younger than 50 years of age, among whom COPD is less prevalent.
When we used only diagnostic codes to define asthma, the association between asthma and invasive pneumococcal disease in our cohort remained significant (adjusted odds ratio, 3.1; 95 percent confidence interval, 2.2 to 4.5), as compared with the odds ratio of 2.4 (95 percent confidence interval, 1.9 to 3.1) reported in our article. Our definition of persons with asthma included patients given a medical diagnosis as well as those treated for asthma and probably excluded those with milder disease. Misclassification of asthma due to errors in diagnosis would probably be nondifferential with respect to the incidence of invasive pneumococcal disease and would therefore make an association more difficult to detect.
Thomas R. Talbot, M.D., M.P.H.
Tina V. Hartert, M.D., M.P.H.
Marie R. Griffin, M.D., M.P.H.
Vanderbilt University School of Medicine
Nashville, TN 37232
tom.talbot@vanderbilt.edu
Deborah Cotton, M.D., M.P.H.
Boston University School of Medicine
Boston, MA 02118
cottond@bu.edu
References
Talbot TR, Hartert TV, Mitchel E, et al. Asthma as a risk factor for invasive pneumococcal disease. N Engl J Med 2005;352:2082-2090.
Kuo AM, Haukoos JS, Witt MD, Babaie ML, Lewis RJ. Recognition of undiagnosed HIV infection: an evaluation of missed opportunities in a predominately urban minority population. AIDS Patient Care STDS 2005;19:239-246.
Garcia-Leoni ME, Moreno S, Rodeno P, Cercenado E, Vicente T, Bouza E. Pneumococcal pneumonia in adult hospitalized patients infected with the human immunodeficiency virus. Arch Intern Med 1992;152:1808-1812.
Redd SC, Rutherford GW III, Sande MA, et al. The role of human immunodeficiency virus infection in pneumococcal bacteremia in San Francisco residents. J Infect Dis 1990;162:1012-1017.
Nuorti JP, Butler JC, Gelling L, Kool JL, Reingold AL, Vugia DJ. Epidemiologic relation between HIV and invasive pneumococcal disease in San Francisco County, California. Ann Intern Med 2000;132:182-190.
To the Editor: Talbot et al. use a case definition of asthma that relies on medical-record review of the diagnostic codes entered or medication-prescription records, or both. Physical-examination data, pulmonary-function assessment, and laboratory information are not incorporated into the case definition. The decision to use a diagnosis of asthma by a noninvestigator physician as the study case definition may result in the misclassification of study subjects. The authors cite two published reports to support the validity of their choice of case definition.1,2 However, the current investigation does not, in fact, use the same case definition as was used in those studies. Moreover, the cited investigations studied patients with asthma who were over the age of 65 years, whereas the average age in the current study population is 28 years and persons older than 49 years were excluded. The authors should acknowledge the possibility of misclassification of study subjects as a potential limitation of their analysis and cite reports that validate their case definition.
Ware G. Kuschner, M.D.
Veterans Affairs Palo Alto Health Care System
Palo Alto, CA 94304
kuschner@stanford.edu
Robert A. Kuschner, M.D.
906 N. Wayne St.
Arlington, VA 22201
References
Hartert TV, Togias A, Mellen BG, Mitchel EF, Snowden MS, Griffin MR. Underutilization of controller and rescue medications among older adults with asthma requiring hospital care. J Am Geriatr Soc 2000;48:651-657.
Hartert TV, Speroff T, Togias A, et al. Risk factors for recurrent asthma hospital visits and death among a population of indigent older adults with asthma. Ann Allergy Asthma Immunol 2002;89:467-473.
The authors reply: We appreciate the comments of Dr. Cotton regarding undiagnosed HIV infection as a risk factor for invasive pneumococcal disease. We reexamined the Tennessee Medicaid program (TennCare) records of cases of invasive pneumococcal disease for the years after patients had received a diagnosis of invasive pneumococcal disease for health care visits that recorded a coded HIV diagnosis or a prescription for HIV-specific medications. In 9.8 percent of the patients coded as having invasive pneumococcal disease (34 of 347) and 0.4 percent of the controls without a diagnosis coded as HIV or other risk factors for invasive pneumococcal disease (22 of 5541), patients were later given a diagnosis of HIV infection. Of patients with invasive pneumococcal disease who had asthma, 5.9 percent (3 of 51) were later given a diagnosis of HIV infection, as compared with 10.5 percent of those without asthma (31 of 296). We agree with Dr. Cotton that HIV testing should be strongly considered for persons who present with invasive pneumococcal disease.
Drs. Kuschner and Kuschner appropriately noted the limitations of the use of administrative data and questioned our definition of asthma. A diagnosis of asthma is established on the basis of the patient's history; documentation of reversible airflow obstruction by spirometry, methacholine challenge or another challenge test, or both; and the exclusion of any alternative diagnoses.1 Previous studies using Medicaid data revealed that diagnostic coding data have an excellent predictive value for asthma diagnosed with the use of these criteria.2,3 The confirmation of the use of diagnostic codes in a younger cohort (15 to 44 years of age) has also been published.4
Those earlier studies also made it possible to assess the validity of our diagnostic algorithm for asthma, which included the use of asthma-specific medications, on the basis of standardized criteria and chart review conducted in an older adult cohort (65 years of age) (unpublished data). This was a more sensitive, though less specific, definition in the older group, since it also identified subjects with chronic obstructive pulmonary disease (COPD). These findings provided the rationale for limiting our study to those younger than 50 years of age, among whom COPD is less prevalent.
When we used only diagnostic codes to define asthma, the association between asthma and invasive pneumococcal disease in our cohort remained significant (adjusted odds ratio, 3.1; 95 percent confidence interval, 2.2 to 4.5), as compared with the odds ratio of 2.4 (95 percent confidence interval, 1.9 to 3.1) reported in our article. Our definition of persons with asthma included patients given a medical diagnosis as well as those treated for asthma and probably excluded those with milder disease. Misclassification of asthma due to errors in diagnosis would probably be nondifferential with respect to the incidence of invasive pneumococcal disease and would therefore make an association more difficult to detect.
Thomas R. Talbot, M.D., M.P.H.
Tina V. Hartert, M.D., M.P.H.
Marie R. Griffin, M.D., M.P.H.
Vanderbilt University School of Medicine
Nashville, TN 37232
tom.talbot@vanderbilt.edu